BiLSTM-VAE Anomaly Weighted Model for Risk-Graded Mine Water Inrush Early Warning
Source abstract
Abstract
A new cascaded model is proposed to improve the accuracy and early warning capability of predicting mine water inrush accidents. The model sequentially applies a Bidirectional Long Short-Term Memory Network (BiLSTM) and a Variational Autoencoder (VAE) to capture the spatio-temporal dependencies between borehole water level data and water inrush events. First, the BiLSTM predicts borehole water levels, and the prediction errors are analyzed to summarize temporal patterns in water level fluctuations. Then, the VAE identifies anomalies in the predicted results. The spatial correlation between borehole water levels, induced by the cone of depression during water inrush, is quantified to assign weights to each borehole. A weighted comprehensive anomaly score is calculated for final prediction. In actual water inrush cases from Xin’an Coal Mine, the BiLSTM-VAE model triggered high-risk alerts 9 h and 30 min in advance, outperforming the conventional threshold-based method by approximately 6 h. Compared with other models, the BiLSTM-VAE demonstrates better timeliness and higher accuracy with lower false alarm rates in mine water inrush prediction. This framework extends the lead time for implementing safety measures and provides a data-driven approach to early warning systems for mine water inrush.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Geoscience and Mining Technology · Earthquake Detection and Analysis · Seismology and Earthquake Studies
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
College of Geology and Environment, Xi’an University of Science and Technology, Xi’an 710054, China
Hebei State Key Laboratory of Mine Disaster Prevention, North China Institute of Science and Technology, Langfang 065201, China
Department of Management Science and Engineering, Khalifa University, Abu Dhabi P.O. Box 127788, United Arab Emirates
Funding
National Key Research and Development Program of China · Khalifa University Faculty Start-Up (FSU) Fund 2023
References (58)
Show source-provided references
1. Fissuring due to inundation of coal mines and its hydrodynamic implications10.1134/S1028334X0704006X ↗
2. A spatial assessment framework for evaluating flood risk under extreme climates10.1016/j.scitotenv.2015.08.094 ↗
3. Lessons learnt from Indian inundation disasters: An analysis of case studies10.1016/j.ijdrr.2016.10.013 ↗
4. Research status and development trends of mine floor water inrush grade prediction10.1007/s10706-017-0408-4 ↗
5. Discriminant analysis of mine water inrush sources with multi-aquifer based on multivariate statistical analysis10.1007/s12665-021-09450-8 ↗
6. Statistics and analysis of coal mine water damage accidents in China in recent 20 years
7. Cause, countermeasures and solutions of water hazards in coal mines in China
8. Research on the prediction of mine water inrush disasters based on multi-factor spatial game reconstruction10.1007/s40948-024-00761-1 ↗
9. Dong, D., and Zhang, J. (2023). Discrimination methods of mine inrush water source. Water, 15.10.3390/w15183237 ↗
10. Location and flux discrimination of water inrush using its spreading process in underground coal mine10.1016/j.ssci.2019.104566 ↗
11. Sun, D., Hou, X., Yang, T., Zhao, Y., Zhang, P., Yang, B., Liu, Y., and Ma, K. (2022). Calibration of water inrush channel and numerical simulation of water inrush process in coal mine roof. Front. Earth Sci., 10.10.3389/feart.2022.931508 ↗
12. Lian, H., Zhang, Q., Yin, S., Yan, T., Yao, H., Yang, S., Kang, J., Xia, X., Li, Q., and Huang, Y. (2024). Integrating microseismic monitoring for predicting water inrush hazards in coal mines. Water, 16.10.3390/w16081168 ↗
13. Numerical simulation on the risk of roof water inrush in Wuyang Coal Mine10.1016/j.ijmst.2012.03.006 ↗
14. Experimental and numerical simulation study on forced ventilation and dust removal of coal mine heading surface10.1007/s40789-024-00667-z ↗
15. Bazaluk, O., Sadovenko, I., Zahrytsenko, A., Saik, P., Lozynskyi, V., and Dychkovskyi, R. (2021). Forecasting Underground Water Dynamics within the Technogenic Environment of a Mine Field. Case Study. Sustain., 13.10.3390/su13137161 ↗
16. Extracting pulse signals in measurement while drilling using optimum denoising methods based on the ensemble empirical mode decomposition10.1016/S1876-3804(12)60107-4 ↗
17. Scenario analysis of mine water inrush hazard using Bayesian networks10.1016/j.ssci.2016.06.013 ↗
18. Prediction reliability of water inrush through the coal mine floor10.1007/s10230-017-0431-y ↗
19. Risk assessment of floor water inrush in coal mines based on MFIM-TOPSIS variable weight model10.1007/s11771-021-4775-x ↗
20. Hybrid CNN-LSTM and IoT-based coal mine hazards monitoring and prediction system10.1016/j.psep.2021.06.005 ↗
21. Quantitative prediction model and prewarning system of water yield capacity (WYC) from coal seam roof based on deep learning and joint advanced detection10.1016/j.energy.2023.130200 ↗
22. Qiu, H., Zhao, H., Xiang, H., Ou, R., Yi, J., Hu, L., Zhu, H., and Ye, M. (2021). Forecasting the incidence of mumps in Chongqing based on a SARIMA model. BMC Public Health, 21.10.1186/s12889-021-10383-x ↗
23. Pressure prediction study of coal mining working face based on Nadam-LSTM10.1109/ACCESS.2023.3302516 ↗
24. Pollutant specific optimal deep learning and statistical model building for air quality forecasting10.1016/j.envpol.2022.118972 ↗
25. Statistical and deep learning models for reference evapotranspiration time series forecasting: A comparison of accuracy, complexity, and data efficiency10.1016/j.compag.2023.108424 ↗
26. Xian, X.B., Wang, L., Wu, X.H., Tang, X.Q., Zhai, X.P., Yu, R., Qu, L.H., and Ye, M.L. (2023). Comparison of SARIMA model, Holt-winters model and ETS model in predicting the incidence of foodborne disease. BMC Infect. Dis., 23.10.1186/s12879-023-08799-4 ↗
27. A weighted one-class support vector machine10.1016/j.neucom.2015.10.097 ↗
28. Ma, D., Duan, H.Y., Cai, X., Li, Z.H., Li, Q., and Zhang, Q. (2018). A Global Optimization-Based Method for the Prediction of Water Inrush Hazard from Mining Floor. Water, 10.10.3390/w10111618 ↗
29. Zhang, Y., Tang, S.F., and Shi, K. (2022). Risk assessment of coal mine water inrush based on PCA-DBN. Sci. Rep., 12.10.1038/s41598-022-05473-8 ↗
30. Jaksik, R., Szumała, K., Dinh, K.N., and Śmieja, J. (2024). Multiomics-Based Feature Extraction and Selection for the Prediction of Lung Cancer Survival. Int. J. Mol. Sci., 25.10.3390/ijms25073661 ↗
31. Photovoltaic forecast based on hybrid PCA–LSSVM using dimensionality reducted data10.1016/j.neucom.2016.01.104 ↗
32. An evaluation of the mine water inrush based on the deep learning of ISMOTE10.1007/s11069-023-05912-3 ↗
33. Intelligent identification of rock mass structural based on point cloud deep learning technology10.1016/j.conbuildmat.2024.139340 ↗
34. Jing, W., Zhao, Z.Q., Wang, X.H., Wang, Y.Q., Wei, X.X., and You, Z.Q. (2024). Intelligent Identification and Prediction of Roof Deterioration Areas Based on Measurements While Drilling. Sensors, 24.10.3390/s24237421 ↗
35. Intelligent identification of foodborne pathogenic bacteria by self-transfer deep learning and ensemble prediction based on single-cell Raman spectrum10.1016/j.talanta.2024.127268 ↗
36. An integrated deep learning model for intelligent recognition of long-distance natural gas pipeline features10.1016/j.ress.2024.110664 ↗
37. TGNet: Intelligent Identification of Thunderstorm Wind Gusts Using Multimodal Fusion10.1007/s00376-024-3308-6 ↗
38. Enhancing information fusion and feature selection efficiency via the PROMETHEE method for multi-source dynamic decision data sets10.1016/j.knosys.2024.112781 ↗
39. An effective multi-source information fusion method for electronic nose and hyperspectral to identify the spring tea quality at different harvesting periods10.1016/j.measurement.2024.116452 ↗
40. Information fusion for large-scale multi-source data based on the Dempster-Shafer evidence theory10.1016/j.inffus.2024.102754 ↗
41. 10-minute forest early wildfire detection: Fusing multi-type and multi-source information via recursive transformer10.1016/j.neucom.2024.128963 ↗
42. Multi-source multi-modal domain adaptation10.1016/j.inffus.2024.102862 ↗
43. Data Assimilation Informed Model Structure Improvement (DAISI) for Robust Prediction Under Climate Change: Application to 201 Catchments in Southeastern Australia10.1029/2023WR036595 ↗
44. Disentangled Seasonal-Trend representation of improved CEEMD-GRU joint model with entropy-driven reconstruction to forecast significant wave height10.1016/j.renene.2024.120345 ↗
45. An improved multi-scale convolutional neural network with gated recurrent neural network model for protein secondary structure prediction10.1007/s00521-024-09822-8 ↗
46. An improved stacking ensemble learning model for predicting the effect of lattice structure defects on yield stress10.1016/j.compind.2023.103986 ↗
47. Integrating crystal structure and numerical data for predictive models of lithium-ion battery materials: A modified crystal graph convolutional neural networks approach10.1016/j.est.2023.110220 ↗
48. A new concept for complex mining of mineral raw material resources from DTEK coal mines based on sustainable development and ESG strategy10.33271/mining17.01.001 ↗
49. A Deep Learning-Based Data-Driven Approach for Predicting Mining Water Inrush From Coal Seam Floor Using Microseismic Monitoring Data10.1109/TGRS.2023.3300012 ↗
50. Predicting mine water inrush accidents based on water level anomalies of borehole groups using long short-term memory and isolation forest10.1016/j.jhydrol.2022.128813 ↗
51. Classification-enhanced LSTM model for predicting river water levels10.1016/j.jhydrol.2024.132535 ↗
52. Bias correction of significant wave height with LSTM neural networks10.1016/j.oceaneng.2024.120015 ↗
53. Real-time rockburst assessment based on a novel hybrid convolutional long short-term memory network based on microseismic monitoring data10.1016/j.engfailanal.2024.109191 ↗
54. Deep learning-based landslide tsunami run-up prediction from synthetic gage data10.1016/j.apor.2024.104360 ↗
55. Enhanced fault diagnosis and remaining useful life prediction of rolling bearings using a hybrid multilayer perceptron and LSTM network model10.1016/j.aej.2024.12.007 ↗
56. Unsupervised aspect-based summarization using variational autoencoders10.1016/j.eswa.2024.126059 ↗
57. Imbalanced fault diagnosis of a conditional variational auto-encoder with transfer and adversarial structures10.1016/j.aei.2024.103032 ↗
58. Optimization-based stacked machine-learning method for seismic probability and risk assessment of reinforced concrete shear walls10.1016/j.eswa.2024.124897 ↗